Evaluating the Impact of FTAs on FDI: A Text-Based Approach

Transcription

1 Evaluating the Impact of FTAs on FDI: A Text-Based Approach Ali Dadkhah Ciuriak Consulting and Infinity Law Dan Ciuriak Ciuriak Consulting, Centre for International Governance Innovation (CIGI), C.D. Howe Institute and BKP Economic Advisors Abstract: June 13-15, 2018 The literature on the impact of bilateral investment treaties (BITs) on foreign direct investment (FDI) has had difficulty establishing a robust effect of BITs on FDI flows. This conclusion extends to FDI chapters in free trade agreements (FTAs), which feature the same substantive content as BITs. The lack of an empirically identifiable effect may reflect econometric difficulties, a minimal role of BITs/FTAs in corporate FDI decisions, or the lack of liberalizing content in the treaties. We consider the latter issue by mapping the legal commitments in FTAs onto the OECD s Foreign Direct Investment Restrictiveness (FDIR) Index, while drawing on the Mode 3 component of the OECD s Services Trade Restrictiveness Index (STRI) to enable greater granularity of treatment of the measures in FTAs. The modified index, which we label the FDI-RI to distinguish it from the FDIR, has the same broad policy areas and the same weighting scheme as the FDIR but the measures under each horizontal policy area (i.e., policies that apply to FDI in general without distinguishing sectors) that are covered in more detail by the STRI have been disaggregated into measures that correspond to those in the STRI. The FDI-RI thus covers both goods and services on a consistent basis, but with greater detail for goods sectors than available under the FDIR. Importantly, we take into account the impact of FTA measures in improving upon each party s bindings under the General Agreement on Trade in Services (GATS). In principle, empirically quantified NTBs reflect both applied restrictions and uncertainty. We identify uncertainty with water in the bound commitments and construct a composite NTB that reflects applied measures and water. This approach allows a determination of whether and how much FTAs liberalize investment by either reducing applied barriers, increasing certainty, or both. Given measures of tax equivalents for non-tariff measures affecting investment, this approach also allows the calculation of a tax equivalent shock for use in quantitative models. We comment on the impact of FDI liberalization in the Comprehensive and Progressive Trans-Pacific Partnership (CPTPP) Agreement using this approach. Keywords: foreign direct investment, FDI, STRI, GATS commitments, water, uncertainty, FTAs JEL Codes: F13, F14, F68

2 1 Introduction The empirical literature on the impact of bilateral investment treaties (BITs) on foreign direct investment (FDI) has had difficulty establishing a robust effect of BITs on FDI flows (Sachs and Sauvant, 2009). This conclusion logically extends to FDI chapters in free trade agreements (FTAs), given they generally feature the same substantive content as BITs. The lack of an empirically identifiable effect may reflect econometric difficulties in isolating the role of BITs/FTAs, a minimal role of BITs/FTAs in corporate FDI decisions, or the lack of liberalizing content in the treaties. We consider the latter issue by adopting a text-based approach to evaluate the liberalization impact of the FTAs on FDI flows for the full range of goods and services sectors. In particular, we map the commitments in BITs/FTAs onto an index based on the OECD s Foreign Direct Investment Restrictiveness (FDIR) Index, but drawing on the Mode 3 component of the OECD s Services Trade Restrictiveness Index (STRI), and on elements of the World Bank s STRI related to licensing. This modified index, which we label the FDI-RI to distinguish it from the FDIR, enables greater granularity of treatment of the measures included in BITs/FTAs. The FDI- RI has the same broad policy areas and the same broad weighting scheme as the FDIR, but the measures under each horizontal policy area (i.e., policies that apply to FDI in general without distinguishing sectors) that are covered in more detail by the STRI Mode 3/World Bank FDI index have been disaggregated into measures that correspond to those in the STRI. The authors of the OECD STRI anticipate this approach: Assigning a unique weight to each measure gives the flexibility to break the STRIs down in several ways. [ ] The STRIs are also presented according to which mode of supply the restriction applies. (Geloso Grosso et al. 2015) The FDI-RI thus covers both goods and services on a consistent basis, but with greater detail for goods sectors than available under the FDIR. An important feature of the approach in this study is to take account of the impact of BITs/FTAs on uncertainty. Firms make state-contingent decisions and future states are uncertain. Given uncertainty about future states and at least partial irreversibility of investment decisions, the opportunity cost of immediate investment (i.e., the option value of delaying and accumulating additional information) must accordingly be included in the cost of investment. As Dixit and Pindyck (1994) emphasize, hurdle rates for firm investments are substantially higher than the cost of capital. They suggest that the value of real options is thus very significant, implying uncertainty is also a very significant factor inhibiting investment. By the same token, empirically quantified NTBs must reflect both applied restrictions and uncertainty. BITs/FTAs address uncertainty introduced by government policy changes. Lucas (1990) emphasizes the role of political risk in shaping the direction of international capital flows. While direct expropriation is largely a thing of the past, indirect expropriation remains an issue for investors. Alfaro et al. (2005) finds protection from government repudiation of contracts and from risk of expropriation to be highly significant determinants of FDI. Azzimonti and Sarte (2007) document the many documented forms of expropriation, direct and indirect, establishing that these are not rare. Studies approaching the issue from an institutional perspective have reached mixed

3 conclusions. Thomas Waelde argues that it would be a sign of negligent management and counsel if political risk management and investment protection were not planned with the potential of investment-treaty based arbitration in mind (cited in Orr, 2007). Kekic and Sauvant (2007) report that the majority of multinationals surveyed take BITs into account in making an investment decision. Yackee (2010) reports that providers of political risk insurance only inconsistently take BITs into account when making underwriting decisions and the majority of the providers he surveyed did not, in fact, view BITs as relevant to their underwriting decisions. At the same time, his results indicate that some underwriters take BITs into account in assessing political risk. As regards uncertainty, the OECD has undertaken work to provide estimates of the difference between the STRI index, which measures restrictiveness in applied policy, and countries bindings under the GATS. This latter difference the water in the GATS bindings provides a measure of the extent to which existing market access restrictions could be increased without violating WTO commitments (Miroudot and Pertel, 2015). Accordingly, water is a measure of the degree of uncertainty faced by firms making investment decisions. Importantly, under this approach to quantifying uncertainty, both actual restrictions and uncertainty are measured by the same indicators and the same weighting scheme. Following the OECD, we identify uncertainty with water in bound commitments and construct a composite NTB that reflects both applied measures and water. This approach allows a determination of whether and how much FTAs liberalize investment by either reducing applied barriers, reducing uncertainty by improving upon a country s bindings under the General Agreement on Trade in Services (GATS), or both. Following Ciuriak and Lysenko (2016), we assume that the value of reducing water (which is measured on the same scale and using the same indicators as applied restrictions) has half the liberalizing value of reducing an equivalent amount of applied restrictions; accordingly, we assign a weight of 0.5 to the bindings in constructing the composite NTB. tably, reduction of applied barriers to FDI without binding creates water. Accordingly, it is essential to determine whether liberalization commitments are bound by the treaty measures or whether there is no penalty for their withdrawal. We convert percentage changes in our composite NTB to changes in trade cost equivalent by applying the percentage changes to ad valorem equivalents (AVEs) of trade costs estimated by Fontagne et al (2016), making the assumption that 50% of the measured AVEs are actionable and 25% of the actionable portion is covered by the measures itemized in the STRI/GTRI. For the simulations, we use a dynamic version of the Global Trade Analysis Project (GTAP) computable general equilibrium (CGE) model that incorporates FDI by introducing a foreignowned representative firm into each GTAP region-sector. For a description, see Ciuriak, Xiao and Dadkhah (2017). In this model, FDI responds in tandem with domestic investment to changes in rates of return in each region-sector based on tariff and NTB reductions; it also responds independently to reductions in NTBs facing investment, based on elements of the CPTPP text that change a country s FDI restrictiveness. Labour supply is assumed to be fixed but we apply a closure such that the quantity of labour (interpreted as productivity) rises in tandem with increases in real wages, consistent with the understanding that trade liberalization transfers market share

4 from lower-productivity firms that pay lower wages to higher-productivity firms that pay higher wages. This paper s main contribution is to describe the development of the FDI-RI for both goods and services, showing how the available indexes can be integrated to increase the granularity of treatment of FTA commitments, and how the impact of an FTA on both applied restrictions and uncertainty can be integrated into impacts on a composite NTB. We illustrate the method with examples from the Trans-Pacific Partnership (CPTPP). The rest of this paper is structured as follows. The following section describes in detail the construction of the FDI NTB in the services sector. Section 3 describes in detail the construction of the FDI NTB in the goods sectors. Section 4 concludes. 2 FDI NTB for Services Sectors To develop an FTA policy shock for liberalization of FDI for services sectors, we code the text of the FTA against the OECD s STRI Mode 3 provisions to yield estimates of: (a) the degree of liberalization implied by the text (in our terminology, this implies changes to the FDI restrictiveness index, FDI-RI); and (b) the extent to which the text binds current practice, thus squeezing water out of the country s commitments under the GATS (in our terminology, this implies changes to the GATS FDI restrictiveness index or G-FDI-RI). The STRI database covers 22 sectors. To better capture FTA commitments (especially in a positive list FTA), we expand coverage to 36 sectors. These sectors and the mapping to the GTAP sectors are listed in Table 1. Table 1: Services sector mapping to GTAP sectors STRI Sector GTAP Sector GTAP Sector 1. Construction Construction Distribution Trade Storage and warehouse Trade Air Transport Air transport Maritime Transport Sea transport Rail freight transport Transport nec Road Transport Transport nec Courier Transport nec Cargo-handling Transport nec Custom brokerage Transport nec Freight forwarding Transport nec Ground-handling Transport nec Telecommunications Communication Securities Financial services nec Commercial Banking Financial services nec Insurance Insurance Accounting Business services nec Architecture Business services nec Building-cleaning Services Business services nec Computer Business services nec Convention Services Business services nec 54

5 22. Engineering Business services nec Legal Business services nec Interdisciplinary R&D Business services nec Environmental Business services nec Nursing, physios & paramedical Business services nec Packaging Services Business services nec Photographic Services Business services nec Rental & Leasing without Operators Business services nec Urban Planning Business services nec Technical testing & analysis services Business services nec Educational Services PubAdmin/Defence/Health/Educat Health Services PubAdmin/Defence/Health/Educat Broadcasting Recreational and other services Sound Recoding Recreational and other services Motion Pictures Recreational and other services 55 Source: Geloso Grosso, et al. (2015); and the authors. The STRI database contains information on Mode 3 services on market access, national treatment, relevant domestic regulation and administrative procedures. The database records policy measures applied on a most-favoured nation (MFN) basis and does not consider preferential treatment entailed in regional trade agreements. The sources of information for the database are laws and regulation in each country. Each entry is documented by reference to the source. The 34 countries included have verified their data and subsequently the database has been subject to peer review assessing their factual accuracy (Geloso Grosso, et al., 2015). For the newly added sectors, the coverage and definition of measures follow the WTO Services Sectoral Classification List (MTN.GNS/W/120) which is based on the Provisional Central Products Classification (CPC Prov.). This classification is used by most countries to schedule specific commitments under the GATS. For instance, for educational services we have included the following sub-categories to assure relatively full coverage of this services sector: Table 2: Expanding the coverage of the STRI Mode 3 Sectors and Sub-sectors Corresponding CPC 5 EDUCATIONAL SERVICES A. Primary education services 921 B. Secondary education services 922 C. Higher education services 923 D. Adult education 924 E. Other education services 929 Source: WTO Services Sectoral Classification List (MTN.GNS/W/120). In order to provide more comprehensive coverage for the BITs/FTAs, we draw on the World Bank STRI (WB STRI) to capture restrictions related to licensing requirements. In particular, WB STRI are used to disaggregate the OECD STRI in the area of transparency requirements. This is of particular interest under some of the latest negotiated BITs/FTAs under the heading Regulatory Framework Domestic Regulation. These new measures include: License automatic if publicly available criteria fulfilled

6 Discriminatory licensing criteria for foreign and domestic firms License allocation: First come first served License allocation: Competitive tender License allocation: At discretion of authority Licensing decision within a certain period required Informing of reasons for license rejection required Licencing and qualification procedures and formalities do not restrict supply of the service Licencing and qualification procedures/formalities do not complicate or delay provision of the service Licencing fees have to be reasonable Decisions of the competent authorities in the licencing or authorisation process are impartial with respect to all applicants Authority is obliged to reach its decision in an independent manner Reasonable period for the submission of an application Establish beforehand the normal timeframe for licensing processing Possibility of re-submitting the application is also guaranteed. These measures are added to the policy area regulatory transparency and administrative requirements. To maintain the overall weight of the FDI-RI, the overall weight for this policy area remains the same and the contribution of the original measures is reduced to accommodate the new measures. Table 3 shows how we have develop the FDI-RI in construction services. Table 3: Development of the construction services FDI-RI Category Measure ID Code Libelle Weight Mode CO _1_1 Maximum foreign equity share allowed (%) CO _1_3 There are limits to the proportion of shares that can be acquired by foreign investors publicly-controlled firms CO _2_1 Legal form: Joint ventures required CO _2_3 Legal form: Foreign branches are prohibited CO _2_11 Legal form: other restrictions CO _4_1 Board of directors: majority must be nationals CO _4_2 Board of directors: majority must be residents CO _4_3 Board of directors: at least one must be national CO _4_4 Board of directors: at least one must be resident CO _4_5 Manager must be national CO _4_6 Manager must be resident CO _5_1 Screening explicitly considers economic interests CO _5_2 Screening exists without exclusion of economic interests CO _5_3 Memo: thresholds for screening projects 0 3 CO _7_1 Acquisition of land and real estate by foreigners is restricted CO _8_1 Restrictions on the type of shares or bonds held by foreign investors CO _9_1 Conditions on subsequent transfer of capital and investments CO _10_1 Restrictions on cross-border mergers and acquisitions CO _12_1 Performance requirements CO _13_1 Discriminatory qualification requirements for building permits to undertake construction work CO _50_1 Other restrictions on foreign entry CO _1_1 Decisions by the regulatory body can be appealed CO _2_1 Firms have redress when business practices restrict competition in a given market CO _50_2 License criteria publicly available CO _50_3 License automatic if publicly available criteria fulfilled

7 CO _50_4 Different licensing criteria for foreign and domestic firms CO _50_5 License allocation: First come first served CO _50_6 License allocation: Competitive tender CO _50_7 License allocation: At discretion of authority CO _50_8 Licensing decision within a certain period required CO _50_9 Informing of reasons for license rejection required CO _50_10 Licensing and qualification procedures and formalities are not restriction for the supply of the service CO _50_11 Licensing and qualification procedures and formalities will not complicate or delay the provision of the service CO _50_12 Licensing fees have to be reasonable CO _50_13 decisions of the competent authorities in the Licensing or authorisation process are impartial with respect to all applicants CO _50_14 Authority is obliged to reach its decision in an independent manner CO _50_15 reasonable period for the submission of an application CO _50_16 establish beforehand the normal timeframe for licensing processing CO _50_17 Possibility of re-submitting the application is also guaranteed Source: the authors. The evaluated reduction to barriers to FDI is based on the impact of the mandatory measures in the BITs/FTAs. Thus, best endeavour clauses are not included in policy shocks. In terms of approach, we first create a template based on the BIT/FTA commitments. In developing the template, each article of the BIT/FTA is mapped to a measure in the FDI-RI. The template is applied to each country in light of its schedule of commitments for establishment. If the FTA changes the answer to not restrictive, the contribution of that measure to the restrictiveness of the regime, which is based on its index weight, is removed. The OECD s STRI Simulator then recalculates the final score for the specific sector. The percentage difference is taken as the degree of reduction of the existing barriers to FDI. Below, we describe the approach to evaluating the impact of an FTA on applied measures, the impact on bindings, the combination of these two effects into impacts on a composite NTB, and the conversion of the index-based change in restrictiveness to a tax equivalent. 2.1 Measuring liberalization of applied measures The methodology covers both horizontal and sector-specific commitments. The horizontal commitments stipulate conditions and restrictions that apply to all sectors of the economy. These measures include investment screening, limitations on board members and managers of firms, impediments on acquiring land and real estate, and so forth. The specific commitments apply to a specific sector as indicated in the schedule. For instance, interest rate regulations pertain only to financial services. Some restrictions tend to impact some sectors more than other; for example, restrictions in public procurement have a particularly large impact on the construction sector in light of the importance of government demand for these services. To measure the liberalization under the BITs/FTAs for FDI, we isolate those measures that affect Mode 3 (FDI). These measures constitute a sub-index that can be referred to as the FDI Restrictiveness Index (FDI-RI in our terminology). The corresponding sub-index for bindings is

8 the GATS FDI Restrictiveness Index or G-FDI-RI. As an example, Table 4 below shows the FDI- RI shock for telecommunication services in Vietnam 1 under the CPTPP. Table 4: FDI-RI shock for telecommunication services in Vietnam Before CPTPP After CPTPP Code Libelle Weight Mode CPTPP FDI-RI G-FDI-RI FDI-RI G-FDI-RI 1_1_1 Maximum foreign equity share Chapter 8, Article allowed (%) (fixed) 1_1_1 Maximum foreign equity share Chapter 8, Section 3, allowed (%) (mobile) Article 2 1_1_1 Maximum foreign equity share Chapter 8, Article allowed (%) (internet) 1_1_3 There are limits to the proportion of shares that can be acquired by foreign investors publicly-controlled firms 1_2_1 Legal form: Joint ventures required Chapter 8, Article (fixed) 1_2_1 Legal form: Joint ventures required Chapter 8, Article (mobile) 1_2_1 Legal form: Joint ventures required Chapter 8, Article (internet) 1_2_11 Legal form: other restrictions Chapter 8, Article _4_1 Board of directors: majority must be Chapter 8, Article nationals 1_4_2 Board of directors: majority must be Chapter 8, Article residents 1_4_3 Board of directors: at least one must Chapter 8, Article be national 1_4_4 Board of directors: at least one must Chapter 8, Article be resident 1_4_5 Manager must be national Chapter 8, Article _4_6 Manager must be resident Chapter 8, Article _5_1 1_5_1 1_5_1 1_5_2 1_5_2 1_5_2 1_5_3 1_5_3 1_5_3 1_7_1 1_8_1 1_8_1 Screening explicitly considers economic interests (fixed) Screening explicitly considers economic interests (mobile) Screening explicitly considers economic interests (internet) Screening exists without exclusion of economic interests (fixed) Screening exists without exclusion of economic interests (mobile) Screening exists without exclusion of economic interests (internet) Memo: thresholds for screening projects (fixed) Memo: thresholds for screening projects (mobile) Memo: thresholds for screening projects (internet) Acquisition and use of land and real estate by foreigners is prohibited or restricted Restrictions on the type of shares or bonds held by foreign investors (fixed) Restrictions on the type of shares or bonds held by foreign investors (mobile) Chapter 8, Article Chapter 8, Article Vietnam s STRI was developed by the study team for the analysis of Vietnam's FDI-RI in a CPTPP/RCEP context

9 1_8_1 Restrictions on the type of shares or Chapter 8, Article bonds held by foreign investors (internet) 1_9_1 Conditions on subsequent transfer of Chapter 8, Article capital and investments 1_10_1 Restrictions on cross-border mergers and acquisitions (fixed) 1_10_1 Restrictions on cross-border mergers and acquisitions (mobile) 1_10_1 Restrictions on cross-border mergers and acquisitions (internet) 1_12_1 Performance requirements (fixed) Chapter 8, Article _12_1 Performance requirements (mobile) Chapter 8, Article _12_1 Performance requirements (internet) Chapter 8, Article _50_1 Other restrictions to movement of people 3_1_1 Foreign suppliers are treated less favourably regarding taxes and eligibility to subsidies 4_1_2 Decisions by the regulatory body Chapter 18, Article can be appealed 4_2_1 Firms have redress when business Chapter 18, Article practices restrict competition in a given market 4_3_1 The government controls at least one major firm in the sector 4_4_1 Publicly-controlled firms are exempted from the application of the general competition law 4_3_2 The government has special voting rights (e.g. golden shares) in any firms 4_6_1 Minimum capital requirements _7_1 Restrictions on advertising _14_1 Unbundling is required Chapter 18: Unbundled Network Elements 4_23_1 Vertical separation is required Chapter 18: (fixed) Regulatory authority 4_23_1 Vertical separation is required Chapter 18: (mobile) Regulatory authority 4_23_1 Vertical separation is required Chapter 18: (internet) Regulatory authority 5_4_1 Time to complete all official procedures required to register a company (in calendar days) 5_5_1 Total cost to complete all official procedures required to register a company (in % of income per capita) 5_6_1 Number of official procedures required to register a company _6_2 5_6_3 5_6_4 5_6_5 License automatic if publicly available criteria fulfilled Different licensing criteria for foreign and domestic firms License allocation: First come first served License allocation: Competitive tender Chapter 5: Conditions for licencing and qualification Chapter 5: Licencing and qualification procedures Chapter 5: Licencing and qualification procedures Chapter 5: Licencing and qualification procedures

10 5_6_6 5_6_7 5_6_8 5_6_9 5_6_10 License allocation: At discretion of authority Licensing decision within a certain period required Informing of reasons for license rejection required Licencing and qualification procedures and formalities are not restriction for the supply of the service Licencing and qualification procedures and formalities will not complicate or delay the provision of the service Chapter 5: Licencing and qualification procedures Chapter 5: Conditions for licencing and qualification Chapter 5: Conditions for licencing and qualification _6_11 licencing fees have to be reasonable _6_12 5_6_13 5_6_14 5_6_15 5_6_16 decisions of the competent authorities in the licencing or authorisation process are impartial with respect to all applicants Authority is obliged to reach its decision in an independent manner reasonable period for the submission of an application establish beforehand the normal timeframe for licensing processing Possibility of re-submitting the application is also guaranteed. Source: the authors Chapter 5: Conditions for licencing and qualification Chapter 18: Regulatory authority Chapter 5: Conditions for licencing and qualification Chapter 5: Conditions for licencing and qualification Total To illustrate how the coding method works, we consider an example from the above table. According to Article Unbundling of Network Elements by Major Suppliers of the CPTPP under the chapter on Telecommunications Networks and Services: Each Party shall provide its telecommunications regulatory body or another appropriate body with the authority to require a major supplier in its territory to offer to public telecommunications service suppliers access to network elements on an unbundled basis on terms and conditions, and at cost-oriented rates, that are reasonable, non-discriminatory and transparent for the supply of public telecommunications services. Each Party may determine the network elements required to be made available in its territory, and the suppliers that may obtain those elements, in accordance with its laws and regulations. Article 43 of the Ordinance PL-UBTVQH10, issued in May 2002, (the Ordinance ) stipulates that all telecommunications enterprises have the right to link their own network to those of other telecommunications enterprises and shall be obliged to allow those other telecommunications enterprises to link and access their own networks or services subject to fair and reasonable conditions. Access and interconnection at every technically and economically feasible point is mandatory for service providers that hold essential equipment and facilities.

11 However, the Ordinance does not define the meaning of essential equipment and facilities, and unbundling of interconnection services is not required. Following the CPTPP, Vietnam has made commitments to provide unbundled network elements to meet reasonable requests for access and use of specific network elements on an unbundled basis subject to a set of principles (such as non-discrimination, transparency, reasonableness of the terms and conditions). Therefore, the CPTPP removes the restriction reflected under the measure 4_14_1: Unbundling is required for the applied level of practice in Vietnam. 2.2 Calculating the Impact on Water To evaluate the impact of improved bindings under a BIT/FTA, we compare the G-FDI-RI and the FDI-RI as bound under the BIT/FTA under examination. The calculation of the binding shock is straightforward: the difference between the G-FDI-RI and the FDI-RI represents water in the bindings. The difference between water pre- and post-fta constitutes the binding shock. te that water can increase under the FTA if applied measures are liberalized but only on an unbound basis. A good example of this is provided by the CPTPP, which contains for some parties a general reservation through a clause in the n-conforming Measures Annex to assure there is no binding beyond the previously GATS negotiated commitments. The [Party] reserves the right to adopt or maintain any measure that is not inconsistent with the [Party s] obligations under Article XVI of the General Agreement on Trade in Services as set but in the [Party s] Schedule of Specific Commitments under the GATS. By allowing the party to revert from the level committed under the CPTPP to as low as the GATS level, the clause creates a level of uncertainty for foreign firms. 2.3 Combining the measures impacting applied restrictions and changing binding To combine the reduction in actual barriers and the reduction of water (the difference between bound and applied market access) into changes to a composite NTB, we follow Ciuriak and Lysenko (2016), who assign reductions in water half the weight of reductions in applied measures. The weighting scheme can be illustrated as follows. Suppose that the initial FDI-RI is 0.4 and the initial G-FDI-RI is 0.8, implying that initial water is 0.4. If both the FDI-RI and water are eliminated by an FTA, the combined effect would be the same as eliminating an NTB index of 0.6 = *0.4 (the value of the FDI-RI plus half the value of water). To illustrate the effect of creating water by liberalizing on an unbound basis, in the above example, if the FDI-RI goes to zero through unbound liberalization, water would rise from 0.4 to 0.8. The unbound nature of the liberalization would thus result in a remaining NTB valued at half the value of resulting value of water, or equal to 0.4. The effective liberalization in this case would be from the initial aggregate NTB index level of 0.6 to one of Calibrating the impact on tax equivalents for FDI The level of observed NTBs on FDI varies widely across sectors and countries. Changes in applied restrictions and the reduction of uncertainty through improved bindings, as measured by

12 percentage changes in the FDI-RI and G-FDI-RI described above, must accordingly be applied to different tax equivalents in different region/sectors. To implement this final step in the calculation of a liberalization shock, the percentage change in the aggregate NTB for FDI into a given region-services sector is applied to tax equivalent. In applications in the GTAP-FDI model (Ciuriak and Xiao, 2014), the tax equivalent is the phantom tax developed as part of that model s framework. Table 5: Sample Calculation FDI-RI G-FDI-RI Water NTB Current Post-BIT/FTA Liberalization % change % Tax equivalent 5% Tax Equivalent Reduction -3.33% Source: the authors. 3 FDI NTB for Goods Sectors As noted, we construct an elaborated FDI-RI for goods sectors by drawing on the horizontal components of the OECD STRI s Mode 3 components and on the World Bank s STRI. We label this elaborated index the FDI-RI to distinguish it from the FDIR. We then consider how the FTA impacts on this index and calculate the percentage change in the index implied by imposing CPTPP restrictions. This percentage change is then applied to the NTB impacting FDI in goods sectors. 3.1 The FDIR Index The FDIR measures statutory restrictions on foreign direct investment in 58 countries, including all OECD and G20 countries, and covers 22 sectors. The FDIR gauges the restrictiveness of a country s FDI rules by looking at the four main types of restrictions on FDI: Foreign equity limitations Screening or approval mechanisms Restrictions on the employment of foreigners as key personnel Operational restrictions, e.g. restrictions on branching and on capital repatriation or on land ownership Good sector FDI restrictions tend to arise mostly in primary sectors such as mining, fishing and agriculture. In contrast to the STRI, where the weight assigned to specific policies varies across different services sector based on expert opinion, the FDIR applies an equal weights scheme (e.g., the presence of a restriction on foreign personnel has the same weight in mining as in manufacturing, etc.) In STRI, the contribution of each policy area to the score in every individual sector is identified; however, in the published FDIR, the contribution of the individual policies within the four broad

13 policy areas within each sector is not identified, in part it would appear because judgement is used in constructing the scores. 2 Table 6 below describes the weighting scheme of the FDIR. As noted, this same scheme applies to each sector. 2 Currently, the OECD is compiling FDI statistics based on some new standards, which will result in significant changes in FDI statistics, including new measures of FDI at the global level. The OECD FDI Statistics database featuring the new statistics will be available in Q In the new database the aggregate statistics are reported based on what is called the asset/liability instead of the current practice which is the directional basis. According to the OECD, this change currently makes it difficult to interpret the statistics due to breaks in series and to different measures being used by different countries during the transition. The extent to which the change in the measurement of FDI will impact on the construction of measures of restriction (at least in terms of expert opinion on the appropriate weights to assign to specific measures) remains to be seen.

14 Table 6: OECD FDIR Weighting Scheme I: Foreign equity limits Weight Start-ups and acquisitions foreign equity allowed 1 Foreign equity < 50% of total equity 0.5 Foreign equity > 50% but < 100% of total equity 0.25 Acquisitions foreign equity allowed 0.5 Foreign equity < 50% of total equity 0.25 Foreign equity > 50% but < 100% of total equity II: Screening and approval Approval required for new FDI/acquisitions of < USD 100mn or < 50% of total equity 0.2 Approval required for new FDI/acquisitions above USD100mn or > 50% of total equity 0.1 tification with discretionary element III: Restrictions on key foreign personnel/directors Foreign key personnel not permitted 0.1 Economic needs test for employment of foreign key personnel 0.05 Time bound limit on employment of foreign key personnel Nationality/residence requirements for board of directors Majority must be nationals At least one must be national 0.02 IV: Other restrictions Establishment of branches not allowed/local incorporation required 0.05 Reciprocity requirement 0.1 Restrictions on profit/capital repatriation Access to local finance 0.05 Acquisition of land for business purposes 3/ 0.1 Land ownership not permitted but leases possible TOTAL Up to 1 Source: OECD. 3.2 Constructing the FDI-RI and the FTA FDI Liberalization Shock In order to construct the shock file for FDI in goods sectors, we construct a general FDI-RI template and apply it to each country in light of its reservations. Table 7 below provides an example based on the CPTPP. The first column shows these measures under each policy area which they correspond to the similar ones in the OECD index. The next column shows how the CPTPP applies to this template. If CPTPP imposes restrictions, we code this as no (e.g., no for maximum foreign equity shares means no such limits may be applied). Table 7: CPTPP Impact on FDI-RI Weight CPTPP Chapter Foreign equity limits Maximum foreign equity share (%) 0.5 Chap. 10, Art. Maximum foreign ownership in local investment companies (%) Chap. 10, Art. Equity restrictions apply to non-locally licensed professionals/firms Chap. 10, Art. Restrictions on cross-border mergers and acquisitions There are restrictions on establishing foreign subsidiaries Chap. 10, Art. CPTPP

15 Foreign companies are prohibited from establishing subsidiaries Chap. 10, Art. Restrictions on the type of shares or bonds held by foreign investors Chap. 10, Art. Screening and approval Approval required for new FDI/acquisitions of < USD 100mn or if 0.2 Chap. 10, Art. corresponding to < 50% of total equity Approval required for new FDI/acquisitions above USD100mn or if 0.1 Chap. 10, Art. corresponding to > 50% of total equity tification with discretionary element Chap. 10, Art. Restrictions on key foreign personnel/directors Board of directors: majority must be nationals Chap. 10, Art. Board of directors: majority must be residents Chap. 10, Art. Board of directors: at least one must be national Chap. 10, Art. Board of directors: at least one must be resident Chap. 10, Art. Manager must be national Chap. 10, Art. Manager must be resident Chap. 10, Art. Other restrictions Local incorporation required 0.05 Restrictions on legal form Chap. 10, Art. Number of firms restricted by quotas Chap. 10, Art. Establishment of foreign firms restricted by economic needs tests Chap. 10, Art. Acquisition of land and real estate by foreigners is prohibited 0.1 Land ownership not permitted but leases possible 0.01 Performance Requirements (Local content requirements) Chap. 9, Art. 10 National Treatment: Foreign suppliers are treated less favourably regarding 0.05 taxes and subsidies National Treatment: n-discriminatory treatment required (other than 0.05 Chap. 10, Art. subsidies) Source: the authors. This is a two- step analysis. In the first step, we look at the text of the CPTPP to see if there are any commitments which could be mapped to the FDI-RI. If this is the case, the answer to the measure in accordance with the text of the CPTPP is added to the column CPTPP. If CPTPP does not affect that measure, the answer is left blank. In the second step, we look at the schedule of reservations to see whether there is any reservation or exclusions in place. If there is a reservation, then we will not remove the previous restrictions in place and keep them the same way.

16 4 Summary and Conclusions This paper develops an elaborated version of the OECD s FDIR, namely the FDI-RI and shows how this index can be used to evaluate the liberalization of FDI under an FTA, taking into account the impact of the FTA on applied restrictions and on uncertainty, where the latter is measured as water in WTO commitments. To translate the text of BITs or FTAs into changes in the FDI-RI, we proceed as follows. First, a general template for the BIT/FTA is created. Second, we apply the template to each individual party in light of the reservations in its schedule of commitments. The percentage change in the FDI-RI implied by an FTA can then be applied to empirical estimates of NTBs facing FDI in goods and services sectors in a given economy, including in aggregations employed in a computable general equilibrium (CGE) modelling context in the Global Trade Analysis Project (GTAP) database, or as independent variables in gravity equations developed to establish the empirical link between the constructed NTB and FDI flows. The major assumption underlying this approach is that BITs/FTAs actually do have a positive impact on FDI inflows. While the empirical literature remains inconclusive on this point, as noted, it seems unlikely that improved legal protections for FDI would not be taken into account when multinational firms make investment decisions and survey evidence discussed above indicates that at least some do. The weak relationship identified in the quantitative impact assessments may reflect econometric issues e.g. the use of dummy variables which fail to distinguish whether individual BITs or FTA investment chapters actually liberalize or introduce significant reductions of uncertainty through improved bindings. This remains an area where further research is required. Other issues that require urgent attention include the development of (a) tax equivalents consistent with the constructed NTBs; and (b) elasticities empirically related to the tax and NTB estimates to mediate the impact of changes in the NTB occasioned by a BIT or FTA. Overall, however, this appears to be a promising direction for the quantitative assessment of BITs or FTA investment chapters on FDI flows. References Alfaro, Laura, Sebnem Kalemli-Ozcan, and Vadym Volosovych Why doesn t from Rich to Poor Countries? An Empirical Investigation, NBER Working Paper Azzimonti, Marina and Pierre-Daniel G. Sarte Barriers to Foreign Direct Investment under Political Instability, Federal Reserve Bank of Richmond Economic Quarterly 93(3), Summer: Ciuriak, Dan and Jingliang Xiao The Trans-Pacific Partnership: Evaluating the 'Landing Zone' for Negotiations. Ciuriak Consulting Working Paper/

17 Ciuriak, Dan and Dmitry Lysenko The Effect of Binding Commitments on Services Trade, C.D. Howe Institute Technical Paper for: Better in than Out? Canada and the Trans- Pacific Partnership, Dixit, Avinash K. and Robert S. Pindyck Investment under Uncertainty. Princeton, NJ: Princeton University Press. Geloso Grosso, M., et al. (2015), "Services Trade Restrictiveness Index (STRI): Scoring and Weighting Methodology", OECD Trade Policy Papers,. 177, OECD Publishing, Paris. DOI: Kekic, L. and Sauvant, K.P. (2007). World Investment Prospects to Foreign Direct Investment and the Challenge of Political Risk. London and New York. The Economist Intelligence Unit and the Columbia Program on International Investment. Lucas Jr., Robert E Why Doesn't Capital Flow from Rich to Poor Countries? American Economic Review 80(2), May: Miroudot, Sébastien and Kätlin Pertel Water in the Gats: Methodology and Results, OECD Trade Policy Papers,. 185, OECD Publishing, Paris. Orr, Ryan J The Impact of BITs on FDI: Do Investors w Ignore BITs? Infrastructure Investment & Dispute Management TDM 2 Sauvant, Karl P. and Lisa E. Sachs (eds) The Effect of Treaties on Foreign Direct Investment Bilateral Investment Treaties, Double Taxation Treaties, and Investment Flows. Oxford University Press. UNCTAD The Role of International Investment Agreements in Attracting Foreign Direct Investment to Developing Countries, UNCTAD Series on International Investment Policies for Development Yackee, Jason Webb Do Bilateral Investment Treaties Promote Foreign Direct Investment? Some Hints from Alternative Evidence, Legal Studies Research Paper Series. 1114, University of Wisconsin Law School.